Abstract
Background:
An increasing number of Major League Baseball (MLB) pitchers are receiving ulnar collateral ligament reconstruction (UCLR). However, a substantial proportion of pitchers do not regain their preinjury level of performance. Currently, limited research has investigated the factors associated with return to level (RTL) after UCLR.
Hypotheses:
The authors hypothesized that (1) pitchers whose performance was primarily dependent on fastball effectiveness may face greater challenges in regaining their prior form, (2) those with a history of high workload were more likely to encounter difficulties in recovery, and (3) pitchers who exhibited suboptimal performance before injury may be less likely to achieve a full RTL.
Study Design:
Case-control study; Level of evidence, 3.
Methods:
This retrospective case-control study analyzed 174 MLB pitchers who underwent primary UCLR between 2015 and 2021. Pitchers were grouped by whether they successfully returned to level, which was defined as returning to MLB within 3 seasons after surgery and recording >100 pitches in at least 1 season. Demographics and pitching metrics were collected to identify risk factors associated with failure to return to prior performance level.
Results:
In this study, 132 pitchers (75.9%) were able to RTL. No significant differences were found in fastball metrics between groups. Univariate analysis revealed that players who were unable to RTL had fewer games started and innings pitched and lower total pitch counts were more frequently relievers during the index season (P < .05). Additionally, significantly worse fielding independent pitching, earned run average, and wins above replacement (WAR) were observed in this group (P < .05). Multivariate analysis indicated that lower preinjury WAR was identified as a significant risk factor for failing to RTL, with a cutoff value of WAR <0.15 (area under the curve, 0.771; P < .001) indicating increased risk.
Conclusion:
The study highlighted that players who were unable to RTL were more frequently relievers. Pitching metrics during the index year, including fewer games started, reduced innings pitched, and lower total pitch counts, were linked to failure to RTL. Performance statistics, such as worse performance in fielding independent pitching, earned run average, and WAR during the index year, were linked to an increased risk of being unable to RTL.
Injury to the ulnar collateral ligament (UCL) is one of the most common injuries in athletes participating in overhead sports, with repetitive valgus forces on the throwing elbow placing significant stress on the joint. 2 The incidence of UCL injury continues to rise over time, especially among baseball pitchers, and UCL reconstruction (UCLR), well known as Tommy John surgery, is commonly performed on these high-level athletes.4,6 The elevated incidence may be attributable to the evolving trends of pitchers having a higher fastball (FB) velocity, a greater workload, or an increased percentage of FB usage among all types of pitches, which were considered risk factors of UCL injuries.11,16,18 Although UCLR was widely believed to improve athletic performance with a satisfactory rate of returning to play in 1 year, ranging from 80% to 97%, return to the same level of play was less frequent and took longer, with 67% to 87% of Major League Baseball (MLB) pitchers returning in about 15 months. 22 When compared with the extensive investigation into risk factors for sustaining a UCL injury, relatively little attention has been given to identifying risk factors associated with failure to return to play or return to sport at a competitive level after UCLR. A recent systematic review revealed inconsistent results in reporting risk factors for inability to return to level (RTL) after UCLR, with factors of age, level of professional play before surgery, and workload being commonly reported. 1 In terms of pitching performance, pre- and postoperative performance metrics varied among pitchers in previous literature, although several studies have indicated that poorer performance in the season before UCL injury may be associated with a reduced likelihood of successfully returning to level after undergoing UCLR.9,10,15,17 McKnight et al 17 demonstrated that a higher pitch count and innings played in the presurgery year trended toward predicting better pitching accuracy after UCLR, suggesting that workload and engagement during the index season may influence recovery outcomes. We aimed to identify significant predictors of failure of RTL in pitchers undergoing UCLR and determine optimal cutoff values for these variables.
In this study, we hypothesized that (1) pitchers whose performance was primarily dependent on FB effectiveness may face greater challenges in regaining their prior form, (2) those with a history of a high workload or prolonged pitching demands were more likely to encounter difficulties in recovery, and (3) pitchers who exhibited suboptimal performance before injury may be less likely to achieve a full return to their previous levels of competition.
Methods
Study Cohort
We conducted a retrospective case-control study of MLB pitchers who underwent UCLR between January 1, 2015, and December 31, 2021, based on the availability of Statcast data. Using the official MLB disabled list and website (MLB.com), we compiled a comprehensive dataset of pitchers who had UCLR during this period. Inclusion criteria required pitchers to have thrown at least 100 pitches in the season before surgery, designated as the “index year.” Pitchers who returned to the MLB level within 3 seasons after surgery and threw at least 100 pitches in a single season were classified as the study group, whereas those who did not meet these criteria were considered the control group. Individuals who underwent revision UCLR procedures were excluded. The flowchart of enrollment is presented in Figure 1.
Flow diagram. MLB, Major League Baseball; UCLR, ulnar collateral ligament reconstruction.
Data Collection and Player Demographics
Demographic and anthropometric data were collected, including age at the time of surgery, body mass index, laterality, and pitching role (starter vs reliever). Pitchers were classified as starters if they were designated as such in at least 50% of their appearances during the season; otherwise, they were categorized as relievers. Injury-related variables were also documented, including the date of UCLR and the date of return to play at the MLB level. To qualify as a player able to RTL, the pitcher was required to have thrown >100 pitches in a calendar year at the MLB level after UCLR.
Pitching performance metrics were obtained from the Statcast system. Annual pitching data were recorded for the study and control groups as follows: FB pitch type (4-seam FB, sinker, or cutter), FB velocity, FB spin rate, FB use percentage, strikeout-to-walk ratio, games played, games started, innings played, total ball use, walks and hits per inning pitched (WHIP), fielding independent pitching (FIP), earned run average (ERA), wins above replacement (WAR), and weighted FB runs per 100 pitches (ie, the mean number of runs that a hitter produced against 100 FBs thrown). For pitchers who used multiple FB types, the most frequently used type was selected for classification. These performance metrics were compared between the study and control groups. Risk factors associated with failure to RTL were analyzed by various performance and pitching metrics.
Statistical Analysis
We performed the statistical analysis using SPSS software (Version 27.0; SPSS Inc). Descriptive statistics were calculated for all available data. For continuous variables, we used the independent t test to perform the analysis. Categorical variables were compared with a chi-square test. Multivariate logistic regression analysis was performed to define risk factors associated with the patients who were unable to RTL after receiving UCLR. Possible factors that have been analyzed include age at injury, weight, height, body mass index, pitcher role, laterality, FB velocity, FB spin rate, FB use percentage, strikeout-to-walk ratio, games played, games started, innings played, total ball use, WHIP, FIP, ERA, WAR, and weighted FB runs per 100 pitches. The enter variable selection method was adopted with a significance level of .10. The results were presented as odds ratio with 95% confidence interval. Receiver operating characteristic curve analysis was adopted to analyze the cutoff value for continuous variables at a significance level of .05. Area under curve was calculated to validate objective image fusion evaluation metrics.
Results
Player Characteristics
A total of 174 MLB players were identified for the analyses. The demographics of the patients, including the mean age at the time of injury, body mass index, and laterality, were similar between the groups, except that the role of the pitcher during the index year showed a statistically significant difference (P = .002). Relievers accounted for a higher percentage of pitchers who were unable to RTL (n = 34; 80.95%) in comparison with pitchers able to RTL (n = 71; 53.79%). The mean interval of RTL was also recorded (Table 1).
Patient Demographics of All Primary UCLR Procedures a
RTL, return to level; UCLR, ulnar collateral ligament reconstruction.
Statistically significant.
Player Performance of the Index Year
Of the 174 patients undergoing UCLR, 132 (75.9%) were able to RTL and 42 (24.1%) were not. The player performance data of index year are presented in Table 2. No significant difference was noted in FB pitching data, including mean velocity, spin rate, use percentage among all types of pitches, or weighted FB runs per 100 pitches. In terms of overall pitching workload, pitchers who were unable to RTL demonstrated significantly fewer games started (7.9 vs 1.7; P < .001), reduced innings pitched (62.2 vs 29.6; P < .001), and lower total pitch counts (992.9 vs 512.8; P < .001). Regarding pitching performance metrics, pitchers able to RTL exhibited significantly superior FIP (P = .008), ERA (P = .008), and WAR (P < .001). Although not statistically significant, the pitchers unable to RTL showed a trend toward higher WHIP (1.7 vs 1.4; P = .085).
Comparison of Player Performance of the Index Year Between Pitchers Able and Unable to RTL After Undergoing Primary UCLR a
ERA, earned run average; FIP, fielding independent pitching (adjusted for park and league); RTL, return to level; UCLR, ulnar collateral ligament reconstruction; WAR, wins above replacement; wFB/C, standardized fastball runs above average per 100 pitches; WHIP, walks and hits per inning pitched.
The year before undergoing primary UCLR.
Statistically significant.
Risk Factor for Being Unable to RTL and Optimal Cutoff Values
We performed risk factor analysis using the enter variable selection method, with the significance test for a risk factor entering and remaining at a significance level of .05. Lower WAR (adjusted odds ratio, 0.165; 95% CI, 0.034-0.805; P = .026) was associated with an increased risk of being unable to RTL (Table 3). Receiver operating characteristic curve analysis was utilized to identify a threshold for WAR in players who were unable to RTL after UCLR. The analysis determined that a cutoff WAR value <0.15 (area under curve, 0.771; P < .001) was indicative of suboptimal return to performance (Figure 2).
Multivariate Analysis of Factors for Returning to Level After Undergoing Primary UCLR a
ERA, earned run average; FIP, fielding independent pitching (adjusted for park and league); UCLR, ulnar collateral ligament reconstruction; WAR, wins above replacement; wFB/C, standardized fastball runs above average per 100 pitches; WHIP, walks and hits per inning pitched.
Statistically significant.
ROC curve. Analysis of the ROC curve illustrates the WAR in assessing the risk of being unable to return to level after receiving primary ulnar collateral reconstruction surgery in Major League Baseball pitchers. AUC, area under the curve; ROC, receiver operating characteristic; WAR, wins above replacement.
Discussion
The most significant finding of this study was the overall RTL rate, with 75.9% of pitchers successfully regaining their preinjury performance levels. A greater proportion of relief pitchers were observed among those who did not RTL. Pitching metrics during the index year, such as a lower number of games started, reduced innings pitched, and fewer total pitches thrown, were associated with unsuccessful RTL. Furthermore, pitchers in this group exhibited inferior FIP, ERA, and WAR performance during the index season. In the multivariate analysis, lower preinjury WAR was identified as a significant risk factor for failing to RTL, with a WAR threshold <0.15.
Biomechanical analysis has shown that FBs generate the highest elbow varus torque, loading rate, and cumulative torque among all pitch types, indicating a greater potential risk for elbow injuries, which mainly consisted of UCL tears. 8 Pitchers who rely heavily on their FB may face greater difficulties returning to play and performing at a high level after elbow injuries, and research exploring the relationship between UCL injuries and FB usage has become increasingly prevalent. A recent study reported that UCLR did not significantly influence basic FB metrics, such as velocity or spin rate, in long-term follow-up, and only short-term changes in pitch usage and advanced performance stats were noted after return. 12 Similar findings were observed in a study analyzing 42 MLB pitchers undergoing UCLR with a 2-year follow-up, 5 which revealed no significant decreases in the spin rate of the 4-seam FB, 2-seam FB, or sliders. However, these analyses were based solely on players who successfully returned to play, which limits the generalizability of their findings. In previous studies analyzing pitchers who successfully returned to level, FB velocity was generally maintained, whereas declines were observed in other performance metrics, such as spin rate and pitch movement.7,21 It still remains possible that pitchers who were unable to RTL experienced greater reductions in spin rate and pitch movement. Furthermore, some studies reported contrasting results except basic FB metrics. Peterson et al 19 demonstrated a statistically significant decline in the percentage of FBs thrown in pitchers receiving UCLR, which might indicate that pitchers experienced a decline in their FB effectiveness against hitters, leading to adjustments in their pitch selection strategy. Moreover, as the FB is known to place the greatest biomechanical stress on the elbow, 8 pitchers may consciously reduce its usage after surgery to mitigate the risk of reinjury of the UCL, leading to the observed decline in FB usage. Strong FB performance was not only defined by fundamental metrics, such as velocity and spin rate, but also took other elements into consideration, such as command, movement, and pitch location, which were equally important in determining a pitcher's ability to dominate hitters. A recent study analyzing 56 MLB pitchers who underwent UCLR surgery revealed a worsened FB accuracy for up to 3 years postoperatively. 17 Similar findings by Portney et al 20 showed that decreased horizontal movement on 4-seam FBs and sliders was also seen in pitchers after returning to play from UCLR. These studies have identified notable differences in pitch selection among pitchers after UCLR, as well as decreases in FB movement and accuracy. The postsurgical changes may reflect underlying alterations in pitching mechanics or modifications in pitch release points. Additionally, a decline in FB effectiveness may contribute to those observed adjustments, suggesting that surgical recovery can influence biomechanical performance and pitching strategy. In our study, we failed to detect the association between FB basic metrics and RTL, and further research is needed to explore changes in FB pitching mechanics among the pitchers receiving the UCLR procedure.
Our study found that relievers had a lower likelihood of successfully returning to their prior performance levels as compared with starters. The observed differences between relievers and starters are likely multifactorial. These findings may be better understood in the context of role-specific demands and the emphasis on velocity that characterizes relief pitching. Unlike starting pitchers, who typically begin innings with no runners on base, relievers often enter games with runners already on base, sometimes in scoring position, thereby experiencing greater situational and psychological stress. Moreover, prior research has demonstrated that relievers generally throw at higher mean velocities as compared with starters (92.0 vs 91.2 mph), 14 reflecting the velocity-oriented nature of their role. The combination of situational demands and emphasis on maximal velocity may contribute to increased biomechanical stresses on the elbow for the relievers and potentially influence performance and recovery trajectories after UCLR. In addition to the role-specific differences, our risk factor analysis identified a lower preinjury WAR as a potential marker associated with an inability to RTL after UCLR. However, this finding should not be interpreted as a contraindication to surgery. It may represent an indicator of a pitcher's baseline performance capacity and overall durability, which could influence postoperative outcomes. Collectively, these findings underscored the significance of preinjury performance in informing postoperative expectations and predicting recovery trajectories. Further research is warranted to clarify how preinjury performance metrics such as WAR may interact with other physiologic and workload factors to determine postoperative success.
A previous study highlighted the finding that despite high rates of returning to play after UCL surgery, only a small percentage of MLB pitchers fully regained their preinjury performance across all key metrics within 3 seasons, with starting pitchers, older age, and higher preoperative velocity being strong negative predictors of full return to form. 15 This indicated that even though players returned to the field, their performance often did not match previous levels. It also underscored the lack of a clear definition for “RTL,” which failed to capture the reality that many players, while technically achieving RTL, did not fully regain their prior effectiveness or value on the mound. Makhni et al 13 reported that pitchers undergoing UCLR were associated with suboptimal pitching performance, including worsened FB velocity, ERA, and WHIP, despite successfully returning to play. A recent systematic review discussing current definitions of return to play indicated that the literature on return to play and RTL after UCL injuries among professional baseball players across all positions remained vague and heterogeneous, with inconsistencies in definitions and methodology that hindered cross-study comparisons. 3 As a result, existing return-to-play or RTL criteria remain insufficient for accurately capturing postinjury outcomes of UCL tear.
Limitations
This study had several limitations. First, as a retrospective study, it did not account for the specific nature of the UCL injuries, the surgical techniques used, the type of graft selected, or the details of postoperative rehabilitation programs. Second, the study analyzed the data when players played at the MLB level, excluding those who may have returned to play at other competitive levels. Last, the data window included the 2020 MLB season, which was notably affected by the COVID-19 pandemic. The shortened season and irregular scheduling likely affected the recovery timelines and performance adjustments of pitchers returning from UCLR, potentially confounding outcome assessments during that period.
Conclusion
The study analyzed 174 MLB pitchers who underwent UCLR, identifying key factors associated with the success to RTL. The results highlighted that a significantly higher proportion of relief pitchers were found in the group unable to RTL, and pitching data during the index year, such as fewer games started, innings pitched, and total pitches thrown, were associated with unsuccessfully returning to level. Performance statistics, including worse FIP, ERA, and WAR during the index year, were linked to an increased risk of being unable to RTL. Risk factor analysis revealed that lower WAR was significantly related to a greater likelihood of failing to RTL.
Footnotes
Final revision submitted August 5, 2025; accepted February 28, 2026.
The authors declared that they have no conflicts of interest in the authorship and publication of this contribution. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.
